PROJECT SUMMARY This U01 is in response to FOA PAR-16-105 Cancer Tissue Engineering Collaborative: Enabling Biomimetic Tissue-Engineered Technologies for Cancer Research. In it we will address the substantial gap in our knowledge of the initiation and progression of metastatic colorectal cancer (CRC). This gap is due, in large part, to the fact that current in vitro, ex vivo, and even in vivo research models are hard to establish or have limited applicability. Accordingly, our objectives are to show the utility of three innovative models of metastasis (a long-term organotypic model and a model that uses multi-organ microfluidic devices) and a new in vivo model (immunocompetent blastocyst model). Using these models, we will investigate the interactive cellular secretome, differentiation, migration and invasion of the primary and metastatic fibroblast tumor microenvironment in order to identify critical contributions to CRC metastasis. Our hypothesis is: inflammation-associated fibroblasts and altered metabolic conditions promote epithelial metastasis via transcriptional regulation. Based on our recent successful engineering of novel models, we propose 3 Aims. Aim 1. To integrate cytokine-secreting stroma into our organotypic model in order to study the influence of inflammation on cancer stem cells (SCs) and on invasion/migration cell phenotypes. Aim 2. To engineer a liver metastasis model using a microfluidic body-on-a- chip platform to study transcriptome and epigenetic reprogramming of metastatic cells. Aim 3. To validate our in vitro models using in vivo immunoproficient murine metastatic models. In Aim 1, decellularized human colons will be repopulated with normal or cancerous fibroblasts from the stroma, and cancer epithelial cells in the form of cancer SCs and organoids to ask how these manipulations influence invasion and differentiation in this model. To complement the cellular findings, we will identify alterations in the transcriptome using RNA-seq and ATAC- seq. Aim 2 uses the innovative body-on-a chip microfluidic device to test autocrine and paracrine proliferative and secretory responses under conditions that favor oncogenic progression, including hypoxia and low glucose. As in Aim 1, epithelia and fibroblasts in the different milieus will be transcriptionally profiled. Aim 3 uses a new immunocompetent blastocyst model and will undergo the same transcriptional profiling. All 3 models will be linked using comparative, informatics approaches. The approach is innovative as it uses human colon tissues in all Aims and asks about the transcriptional contribution and landscape of accessible chromatin in each model. Our multidisciplinary team includes a colorectal surgeon-scientist and biomedical engineers with expertise in cancer SCs, signaling, metastases, and organotypic and organ-on-a-chip platforms. The findings will be significant, as they will establish improved approaches for studying the pathogenesis of advanced CRC, and for developing patient-derived, high-throughput cell models to test therapeutic interventions.